Ultralow noise new signal processing method and system based on ultrasonic signal feature vector

Abstract

The invention provides an ultra-low noise new signal processing method and system based on an ultrasonic signal feature vector, belongs to the technical field of signal data processing, and solves the problems that an existing ultra-low noise new signal processing method is low in processing efficiency, poor in precision and the like. The invention discloses an ultra-low noise new signal processing method and system based on an ultrasonic signal feature vector. Comprising an ultrasonic data signal acquisition module, a data preprocessing module, a feature vector extraction module, an ultra-low noise removal module, a signal enhancement module, a data compression and reconstruction module, a signal output and optimization module, a post-processing and result analysis module and a deep learning module. According to the invention, ultra-low noise processing can be carried out on an ultra-low noise new signal; performing high-precision feature extraction; efficient data processing; the system is suitable for different types of ultrasonic signals and application scenes, and has high flexibility and adaptability; the detection precision is improved; and the signal readability is enhanced.

Description

Technical Field

[0001] The present invention belongs to the technical field of signal data processing, and relates to a new ultra-low noise signal processing method based on ultrasonic signal characteristic vectors, in particular to a new ultra-low noise signal processing system based on ultrasonic signal characteristic vectors. Background Art

[0002] Ultrasonic testing plays an important role in the field of non-destructive testing and is widely used in many fields such as material defect detection, medical diagnosis, and biological tissue imaging. Ultrasonic signals carry rich information, and by analyzing and processing them, accurate evaluation and diagnosis of the object being tested can be achieved.

[0003] The main challenges faced by ultrasonic detection signals are:

[0004] 1. Noise interference: In practical applications, ultrasonic detection signals are often interfered by various noises, such as base white noise, environmental noise, etc. The presence of these noises will seriously affect the quality of the signal, and then affect the subsequent signal analysis and processing.

[0005] 2. Data storage capacity: Ultrasonic testing usually generates a large amount of data, and the storage and transmission of this data requires a lot of resources and space. How to effectively reduce the amount of data storage while ensuring signal quality is an urgent problem to be solved in the field of ultrasonic signal processing.

[0006] 3. Position accuracy: In ultrasonic testing, the position accuracy of defects is crucial for subsequent diagnosis and repair. Traditional maximum value extraction methods may not meet the needs of high-precision testing in some cases.

[0007] Based on this, we propose a new ultra-low noise signal processing method and system based on ultrasonic signal feature vector, which can perform ultra-low noise processing on ultra-low noise new signals; high-precision feature extraction; efficient data processing; the system is suitable for different types of ultrasonic signals and application scenarios, with strong flexibility and adaptability; improves detection accuracy; and enhances signal readability. Summary of the invention

[0008] The purpose of the present invention is to address the above-mentioned problems in the existing technology and propose a new ultra-low noise signal processing method and system based on ultrasonic signal feature vectors. The technical problems to be solved by the invention are: how to achieve ultra-low noise processing of ultra-low noise new signals; high-precision feature extraction; efficient data processing; the system is suitable for different types of ultrasonic signals and application scenarios, with strong flexibility and adaptability; improve detection accuracy; and enhance signal readability.

[0009] The purpose of the present invention can be achieved by the following technical solutions:

[0010] like Figure 1-Figure 10 As shown, the new ultra-low noise signal processing system based on ultrasonic signal feature vector includes ultrasonic data signal acquisition module, data preprocessing module, feature vector extraction module, ultra-low noise removal module, signal enhancement module, data compression and reconstruction module, signal output and optimization module, post-processing and result analysis module and deep learning module.

[0011] The ultrasonic data signal acquisition module includes an ultrasonic sensor or probe, an analog-to-digital converter and a digital-to-analog converter.

[0012] The data preprocessing module includes a signal amplifier, a filter, a DC offset removal circuit and a signal conditioning circuit.

[0013] The filter includes but is not limited to a spatial domain filter, a frequency domain filter, an adaptive filter, a bandpass filter, and a bandstop filter.

[0014] The feature vector extraction module includes a signal segmentation unit, a feature selection unit, a time domain feature vector extraction unit, a frequency domain feature vector extraction unit and a feature vector construction unit.

[0015] The characteristic vector of the characteristic vector extraction module includes parameters such as arrival time, bandwidth factor, center frequency, amplitude and phase of the echo.

[0016] The time domain feature vector extraction unit extracts feature parameters related to noise and signal including but not limited to peak value, mean value and variance.

[0017] The ultra-low noise removal module includes a noise modeling and estimation unit, a denoising algorithm design unit and a signal denoising unit.

[0018] The specific algorithms of the ultra-low noise removal module include but are not limited to adaptive filtering, Kalman filtering, Wiener filtering, spectral subtraction, wavelet transform denoising, sparse representation and reconstruction, singular value decomposition denoising, neural network denoising, machine learning algorithm, sliding average method and median filtering.

[0019] The signal enhancement module includes a frequency modulator, a signal amplifier and a signal reconstruction unit.

[0020] Specific algorithms of the signal enhancement module include but are not limited to acoustic beamforming algorithms and active noise control algorithms.

[0021] The data compression and reconstruction module includes a data compression unit and a data reconstruction unit.

[0022] The specific algorithms of the data compression unit include but are not limited to lossy compression and lossless compression, and the specific algorithms of the data reconstruction unit include but are not limited to orthogonal matching pursuit algorithm, compressed sampling matching pursuit algorithm and signal interpolation algorithm.

[0023] The signal output and optimization module includes outputting the optimized signal to the subsequent processing system or application equipment as well as the output drive circuit, signal conditioning circuit and output interface; the signal output format is selected according to specific needs, including but not limited to digital signals and analog signals.

[0024] The signal output and optimization module is optimized by, but not limited to, smoothing and artifact removal techniques to optimize the signal, and uses median filtering and mean filtering smoothing techniques to reduce glitches and artifacts in the signal.

[0025] The post-processing and result analysis module includes a post-processing unit and a result analysis unit.

[0026] The method optimization and iteration module includes an algorithm evaluation unit, an algorithm optimization unit, a system integration and testing unit

[0027] The deep learning module learns the inherent variation patterns and characteristics of data by training a large number of data samples, realizes the prediction and classification of new data, and improves the intelligent level of signal processing.

[0028] A new ultra-low noise signal processing method based on ultrasonic signal feature vector includes the following processing steps:

[0029] Step 1: Acquisition of ultrasonic signals: Use a high-performance ultrasonic sensor or probe to transmit ultrasonic signals in the target area, receive the reflected signals, convert the received ultrasonic signals into electrical signals, and perform analog-to-digital conversion for subsequent digital signal processing. Finally, convert the discrete digital signals into continuous analog signals. It is necessary to ensure that the collected signals have sufficient signal-to-noise ratio, clarity and resolution for subsequent processing.

[0030] Step 2, data preprocessing: preliminary processing of the collected digital signal, including but not limited to signal amplification, DC component removal, gain control and filtering, to remove some obvious noise and interference, to improve the signal with a reliable signal-to-noise ratio, and to retain the effective information of the signal;

[0031] Step 3, feature vector extraction: According to the characteristics of ultrasonic signals and application requirements, select feature parameters that can reflect the main characteristics and change laws of the signal; use signal processing technology to extract key information of the characteristics of time domain characteristics and frequency domain characteristics from ultrasonic signals, and then construct feature vectors based on the extracted time domain and frequency domain characteristics. The feature vectors should be able to comprehensively and accurately reflect the main characteristics and change laws of ultrasonic signals, and have certain robustness and anti-interference capabilities;

[0032] Step 4: Ultra-low noise signal processing: Analyze the noise in the ultrasonic signal, determine the type, source and characteristics of the noise, then model the noise, understand the source, type, characteristics and distribution of the noise, and use the noise estimation method to accurately estimate the noise; design a specific filtering algorithm or denoising algorithm based on the noise analysis results and feature vectors, and then apply the denoising algorithm to denoise the ultrasonic signal, further reduce the noise level in the signal, improve the signal purity, and retain the main characteristics and useful information of the signal;

[0033] Step 5: Signal enhancement: Based on denoising, the signal is further enhanced and restored through the signal enhancement algorithm to enhance the effective components of the signal, improve the signal-to-noise ratio, and improve the signal's amplitude, strength, clarity, contrast, anti-interference ability, transmission efficiency, detection ability, amplitude and power to ensure the purity and stability of the signal;

[0034] Step 6: Data compression and reconstruction: Data compression is performed on the denoised and enhanced ultrasonic signal to reduce the burden of data storage and transmission, improve the efficiency of data processing, help reduce the operating cost of the system, and improve the overall performance of the system; the data compression algorithm through lossy compression or lossless compression should be able to retain the main features and useful information of the signal while reducing the redundancy and complexity of the data;

[0035] Reconstruct the compressed data according to the feature vector to restore the original information and detailed features of the signal, ensuring that the reconstructed signal is not distorted and maintains high position accuracy; accurately restore the waveform and characteristic parameters of the signal through the data reconstruction algorithm, while avoiding the introduction of new noise and interference, maintaining the accuracy and reliability of the signal; output ultra-low noise ultrasonic signals;

[0036] Step 7: Signal optimization and output: The reconstructed signal is further optimized, including but not limited to smoothing and artifact removal, to reduce burrs and artifacts in the signal, enhance the signal strength through algorithms, improve the signal-to-noise ratio, further reduce noise interference and improve signal accuracy;

[0037] Output the optimized signal to the subsequent processing system or application equipment. The output format can be selected according to specific needs. The output format includes but is not limited to digital signals and analog signals. This step is the end of the signal processing flow. It is necessary to ensure that the quality and stability of the output signal meet the requirements of subsequent processing.

[0038] Step 8, post-processing and result analysis: smooth the processed signal to remove possible artifacts or outliers; perform data normalization or standardization for subsequent analysis and application;

[0039] Perform quality assessment on the processed signal, including but not limited to the degree of improvement in signal-to-noise ratio and degree of distortion; use visualization tools to display and analyze the signal, including but not limited to time domain waveforms and spectrum diagrams;

[0040] Step 9: Method optimization and iteration: Algorithm optimization: Based on actual application scenarios and needs, combined with artificial intelligence and deep learning modules, evaluate the performance of the algorithm, including but not limited to denoising effect, signal enhancement effect and data compression efficiency, and use comparative experiments and statistical analysis methods for continuous evaluation and adjustment;

[0041] Adjust the feature vector extraction method and denoising algorithm design parameters based on the processing results, optimize and iterate the feature extraction, noise estimation and denoising algorithms. The system can adaptively adjust the processing parameters to improve the accuracy, performance and efficiency of signal processing and reduce computational complexity and resource consumption.

[0042] System integration and testing: Integrate the ultra-low noise signal processing method into the ultrasonic signal processing system, perform system testing and verification, and ensure the stability and reliability of the system.

[0043] Compared with the prior art, the new ultra-low noise signal processing method and system based on ultrasonic signal feature vector has the following advantages:

[0044] Ultra-low noise processing: The system can significantly reduce the noise level in ultrasonic signals and achieve ultra-low noise signal processing effects. This helps to improve the accuracy and reliability of the signal, which is especially important when detecting and identifying weak signals.

[0045] High-precision feature extraction: Using advanced machine learning and deep learning algorithms, the system can accurately extract the feature vectors of ultrasonic signals. These feature vectors can reflect the main characteristics and change patterns of the signal, providing strong support for subsequent signal processing.

[0046] Efficient data processing: The system uses efficient data compression and reconstruction algorithms to reduce the burden of data storage and transmission and improve the efficiency of data processing. This helps reduce the operating costs of the system and improve the overall performance of the system.

[0047] Flexibility and adaptability: The system is suitable for different types of ultrasonic signals and application scenarios, and has strong flexibility and adaptability. This enables the system to work in different environments and conditions to meet diverse needs.

[0048] Improve detection accuracy: Since the system can significantly reduce noise interference and enhance signal strength, the precision and accuracy of ultrasonic detection can be improved. This is of great significance in non-destructive testing, medical diagnosis and other fields.

[0049] Enhanced signal readability: After system processing, the noise of the ultrasonic signal is effectively removed, and the signal becomes clearer and easier to interpret. This helps operators to more accurately judge and analyze signal characteristics and improve work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a system flow chart of the present invention.

[0051] Figure 2 It is a structural schematic diagram of ultrasonic signal acquisition in the present invention.

[0052] Figure 3 It is a structural schematic diagram of data preprocessing in the present invention.

[0053] Figure 4 It is a structural schematic diagram of feature vector extraction in the present invention.

[0054] Figure 5 It is a structural schematic diagram of ultra-low noise signal processing in the present invention.

[0055] Figure 6 It is a schematic diagram of the structure of signal enhancement in the present invention.

[0056] Figure 7 It is a structural schematic diagram of data compression and reconstruction in the present invention.

[0057] Figure 8 It is a structural schematic diagram of signal optimization and output in the present invention.

[0058] Fig. 9 It is a structural schematic diagram of post-processing and result analysis in the present invention.

[0059] Fig.10 It is a schematic diagram of the structure of the method optimization and iteration in the present invention. DETAILED DESCRIPTION

[0060] The following are specific embodiments of the present invention and the accompanying drawings to further describe the technical solution of the present invention, but the present invention is not limited to these embodiments.

[0061] like Figure 1-Figure 10 As shown, the new ultra-low noise signal processing system based on ultrasonic signal feature vector includes ultrasonic data signal acquisition module, data preprocessing module, feature vector extraction module, ultra-low noise removal module, signal enhancement module, data compression and reconstruction module, signal output and optimization module, post-processing and result analysis module and deep learning module.

[0062] The ultrasonic data signal acquisition module includes an ultrasonic sensor or probe, an analog-to-digital converter and a digital-to-analog converter.

[0063] The data preprocessing module includes a signal amplifier, a filter, a DC offset removal circuit and a signal conditioning circuit.

[0064] The filters include but are not limited to spatial domain filters, frequency domain filters, adaptive filters, bandpass filters, and bandstop filters.

[0065] The feature vector extraction module includes a signal segmentation unit, a feature selection unit, a time domain feature vector extraction unit, a frequency domain feature vector extraction unit and a feature vector construction unit.

[0066] The feature vector of the feature vector extraction module contains parameters such as arrival time, bandwidth factor, center frequency, amplitude and phase of the echo.

[0067] The time domain feature vector extraction unit extracts feature parameters related to noise and signal including but not limited to peak value, mean value and variance.

[0068] The ultra-low noise removal module includes a noise modeling and estimation unit, a denoising algorithm design unit and a signal denoising unit.

[0069] The specific algorithms of the ultra-low noise removal module include but are not limited to adaptive filtering, Kalman filtering, Wiener filtering, spectral subtraction, wavelet transform denoising, sparse representation and reconstruction, singular value decomposition denoising, neural network denoising, machine learning algorithm, sliding average method and median filtering.

[0070] The signal enhancement module includes a frequency modulator, a signal amplifier and a signal reconstruction unit.

[0071] Specific algorithms of the signal enhancement module include, but are not limited to, acoustic beamforming algorithms and active noise control algorithms.

[0072] The data compression and reconstruction module includes a data compression unit and a data reconstruction unit.

[0073] The specific algorithms of the data compression unit include but are not limited to lossy compression and lossless compression, and the specific algorithms of the data reconstruction unit include but are not limited to orthogonal matching pursuit algorithm, compressed sampling matching pursuit algorithm and signal interpolation algorithm.

[0074] The signal output and optimization module includes outputting the optimized signal to the subsequent processing system or application equipment as well as the output drive circuit, signal conditioning circuit and output interface; the signal output format is selected according to specific needs, including but not limited to digital signals and analog signals.

[0075] The signal output and optimization module is optimized by, but not limited to, smoothing and artifact removal techniques to optimize the signal, and uses median filtering and mean filtering smoothing techniques to reduce glitches and artifacts in the signal.

[0076] The post-processing and result analysis module includes a post-processing unit and a result analysis unit.

[0077] The method optimization and iteration module includes algorithm evaluation unit, algorithm optimization unit, system integration and testing unit

[0078] The deep learning module learns the inherent changing patterns and characteristics of the data by training a large number of data samples, realizes the prediction and classification of new data, and improves the intelligent level of signal processing.

[0079] The new ultra-low noise signal processing method based on ultrasonic signal feature vector includes the following processing steps:

[0080] Step 1: Acquisition of ultrasonic signals: Use a high-performance ultrasonic sensor or probe to transmit ultrasonic signals in the target area, receive the reflected signals, convert the received ultrasonic signals into electrical signals, and perform analog-to-digital conversion for subsequent digital signal processing. Finally, convert the discrete digital signals into continuous analog signals. It is necessary to ensure that the collected signals have sufficient signal-to-noise ratio, clarity and resolution for subsequent processing.

[0081] Step 2, data preprocessing: preliminary processing of the collected digital signal, including but not limited to signal amplification, DC component removal, gain control and filtering, to remove some obvious noise and interference, to improve the signal with a reliable signal-to-noise ratio, and to retain the effective information of the signal;

[0082] Step 3, feature vector extraction: According to the characteristics of ultrasonic signals and application requirements, select feature parameters that can reflect the main characteristics and change laws of the signal; use signal processing technology to extract key information of the characteristics of time domain characteristics and frequency domain characteristics from ultrasonic signals, and then construct feature vectors based on the extracted time domain and frequency domain characteristics. The feature vectors should be able to comprehensively and accurately reflect the main characteristics and change laws of ultrasonic signals, and have certain robustness and anti-interference capabilities;

[0083] Step 4: Ultra-low noise signal processing: Analyze the noise in the ultrasonic signal, determine the type, source and characteristics of the noise, then model the noise, understand the source, type, characteristics and distribution of the noise, and use the noise estimation method to accurately estimate the noise; design a specific filtering algorithm or denoising algorithm based on the noise analysis results and feature vectors, and then apply the denoising algorithm to denoise the ultrasonic signal, further reduce the noise level in the signal, improve the signal purity, and retain the main characteristics and useful information of the signal;

[0084] Step 5: Signal enhancement: Based on denoising, the signal is further enhanced and restored through the signal enhancement algorithm to enhance the effective components of the signal, improve the signal-to-noise ratio, and improve the signal's amplitude, strength, clarity, contrast, anti-interference ability, transmission efficiency, detection ability, amplitude and power to ensure the purity and stability of the signal;

[0085] Step 6: Data compression and reconstruction: Data compression is performed on the denoised and enhanced ultrasonic signal to reduce the burden of data storage and transmission, improve the efficiency of data processing, help reduce the operating cost of the system, and improve the overall performance of the system; the data compression algorithm through lossy compression or lossless compression should be able to retain the main features and useful information of the signal while reducing the redundancy and complexity of the data;

[0086] Reconstruct the compressed data according to the feature vector to restore the original information and detailed features of the signal, ensuring that the reconstructed signal is not distorted and maintains high position accuracy; accurately restore the waveform and characteristic parameters of the signal through the data reconstruction algorithm, while avoiding the introduction of new noise and interference, maintaining the accuracy and reliability of the signal; output ultra-low noise ultrasonic signals;

[0087] Step 7: Signal optimization and output: The reconstructed signal is further optimized, including but not limited to smoothing and artifact removal, to reduce burrs and artifacts in the signal, enhance the signal strength through algorithms, improve the signal-to-noise ratio, further reduce noise interference and improve signal accuracy;

[0088] Output the optimized signal to the subsequent processing system or application equipment. The output format can be selected according to specific needs. The output format includes but is not limited to digital signals and analog signals. This step is the end of the signal processing flow. It is necessary to ensure that the quality and stability of the output signal meet the requirements of subsequent processing.

[0089] Step 8, post-processing and result analysis: smooth the processed signal to remove possible artifacts or outliers; perform data normalization or standardization for subsequent analysis and application;

[0090] Perform quality assessment on the processed signal, including but not limited to the degree of improvement in signal-to-noise ratio and degree of distortion; use visualization tools to display and analyze the signal, including but not limited to time domain waveforms and spectrum diagrams;

[0091] Step 9: Method optimization and iteration: Algorithm optimization: Based on actual application scenarios and needs, combined with artificial intelligence and deep learning modules, evaluate the performance of the algorithm, including but not limited to denoising effect, signal enhancement effect and data compression efficiency, and use comparative experiments and statistical analysis methods for continuous evaluation and adjustment;

[0092] Adjust the feature vector extraction method and denoising algorithm design parameters based on the processing results, optimize and iterate the feature extraction, noise estimation and denoising algorithms. The system can adaptively adjust the processing parameters to improve the accuracy, performance and efficiency of signal processing and reduce computational complexity and resource consumption.

[0093] System integration and testing: Integrate the ultra-low noise signal processing method into the ultrasonic signal processing system, perform system testing and verification, and ensure the stability and reliability of the system.

[0094] In summary, the processing method and system have ultra-low noise processing: the system can significantly reduce the noise level in the ultrasonic signal and achieve an ultra-low noise signal processing effect. This helps to improve the accuracy and reliability of the signal, which is especially important when detecting and identifying weak signals.

[0095] High-precision feature extraction: Using advanced machine learning and deep learning algorithms, the system can accurately extract the feature vectors of ultrasonic signals. These feature vectors can reflect the main characteristics and change patterns of the signal, providing strong support for subsequent signal processing.

[0096] Efficient data processing: The system uses efficient data compression and reconstruction algorithms to reduce the burden of data storage and transmission and improve the efficiency of data processing. This helps reduce the operating costs of the system and improve the overall performance of the system.

[0097] Flexibility and adaptability: The system is suitable for different types of ultrasonic signals and application scenarios, and has strong flexibility and adaptability. This enables the system to work in different environments and conditions to meet diverse needs.

[0098] Improve detection accuracy: Since the system can significantly reduce noise interference and enhance signal strength, the precision and accuracy of ultrasonic detection can be improved. This is of great significance in non-destructive testing, medical diagnosis and other fields.

[0099] Enhanced signal readability: After system processing, the noise of the ultrasonic signal is effectively removed, and the signal becomes clearer and easier to interpret. This helps operators to more accurately judge and analyze signal characteristics and improve work efficiency.

[0100] The specific embodiments described herein are merely examples of the spirit of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in similar ways, but they will not deviate from the spirit of the present invention or exceed the scope defined by the appended claims.

Claims

1. A new ultra-low noise signal processing system based on ultrasonic signal feature vector, characterized in that: It includes ultrasonic data signal acquisition module, data preprocessing module, feature vector extraction module, ultra-low noise removal module, signal enhancement module, data compression and reconstruction module, signal output and optimization module, post-processing and result analysis module and deep learning module.

2. The new ultra-low noise signal processing system based on ultrasonic signal characteristic vector according to claim 1 is characterized in that: The ultrasonic data signal acquisition module includes an ultrasonic sensor or probe, an analog-to-digital converter and a digital-to-analog converter; the data preprocessing module includes a signal amplifier, a filter, a DC offset removal circuit and a signal conditioning circuit; the filter includes but is not limited to a spatial domain filter, a frequency domain filter, an adaptive filter, a bandpass filter, and a bandstop filter.

3. The new ultra-low noise signal processing system based on ultrasonic signal characteristic vector according to claim 2 is characterized in that: The feature vector extraction module includes a signal segmentation unit, a feature selection unit, a time domain feature vector extraction unit, a frequency domain feature vector extraction unit and a feature vector construction unit; the feature vector of the feature vector extraction module contains parameters such as arrival time, bandwidth factor, center frequency, amplitude and phase of echo; the time domain feature vector extraction unit extracts feature parameters related to noise and signal including but not limited to peak value, mean and variance.

4. The new ultra-low noise signal processing system based on ultrasonic signal characteristic vector according to claim 3 is characterized in that: The ultra-low noise removal module includes a noise modeling and estimation unit, a denoising algorithm design unit and a signal denoising unit; the specific algorithms of the ultra-low noise removal module include but are not limited to adaptive filtering, Kalman filtering, Wiener filtering, spectral subtraction, wavelet transform denoising, sparse representation and reconstruction, singular value decomposition denoising, neural network denoising, machine learning algorithm, sliding average method and median filtering.

5. The new ultra-low noise signal processing system based on ultrasonic signal characteristic vector according to claim 4 is characterized in that: The signal enhancement module includes a frequency modulator, a signal amplifier and a signal reconstruction unit; the specific algorithm of the signal enhancement module includes but is not limited to an acoustic beamforming algorithm and an active noise control algorithm.

6. The new ultra-low noise signal processing system based on ultrasonic signal characteristic vector according to claim 5 is characterized in that: The data compression and reconstruction module includes a data compression unit and a data reconstruction unit; the specific algorithms of the data compression unit include but are not limited to lossy compression method and lossless compression method, and the specific algorithms of the data reconstruction unit include but are not limited to orthogonal matching pursuit algorithm, compressed sampling matching pursuit algorithm and signal interpolation algorithm.

7. The new ultra-low noise signal processing system based on ultrasonic signal characteristic vector according to claim 6 is characterized in that: The signal output and optimization module includes outputting the optimized signal to a subsequent processing system or application device as well as an output drive circuit, a signal conditioning circuit and an output interface; the signal output format is selected according to specific needs, including but not limited to digital signals and analog signals; the optimization method of the signal output and optimization module is but not limited to smoothing processing and artifact removal technical methods to optimize the signal, and adopts median filtering and mean filtering smoothing technology to reduce burrs and artifacts in the signal.

8. The new ultra-low noise signal processing system based on ultrasonic signal characteristic vector according to claim 7 is characterized in that: The post-processing and result analysis module includes a post-processing unit and a result analysis unit; the method optimization and iteration module includes an algorithm evaluation unit, an algorithm optimization unit, and a system integration and testing unit.

9. The new ultra-low noise signal processing system based on ultrasonic signal characteristic vector according to claim 8 is characterized in that: The deep learning module learns the inherent variation patterns and characteristics of data by training a large number of data samples, realizes the prediction and classification of new data, and improves the intelligent level of signal processing.

10. A processing method of the new ultra-low noise signal processing system based on ultrasonic signal feature vector as claimed in claims 1 to 9, characterized in that: The following processing steps are included: Step 1, ultrasonic signal acquisition: Use a high-performance ultrasonic sensor or probe to transmit ultrasonic signals in the target area, receive the reflected signals, convert the received ultrasonic signals into electrical signals, and perform analog-to-digital conversion for subsequent digital signal processing. Finally, the discrete digital signals are converted into continuous analog signals. It is necessary to ensure that the collected signals have sufficient signal-to-noise ratio, clarity and resolution for subsequent processing. Step 2, data preprocessing: preliminary processing of the collected digital signal, including but not limited to signal amplification, DC component removal, gain control and filtering, to remove some obvious noise and interference, to improve the signal with a reliable signal-to-noise ratio, and to retain the effective information of the signal; Step 3, feature vector extraction: According to the characteristics of ultrasonic signals and application requirements, select feature parameters that can reflect the main characteristics and change laws of the signal; use signal processing technology to extract key information of the characteristics of time domain characteristics and frequency domain characteristics from ultrasonic signals, and then construct feature vectors based on the extracted time domain and frequency domain characteristics. The feature vectors should be able to comprehensively and accurately reflect the main characteristics and change laws of ultrasonic signals, and have certain robustness and anti-interference capabilities; Step 4: Ultra-low noise signal processing: Analyze the noise in the ultrasonic signal, determine the type, source and characteristics of the noise, then model the noise, understand the source, type, characteristics and distribution of the noise, and use the noise estimation method to accurately estimate the noise; design a specific filtering algorithm or denoising algorithm based on the noise analysis results and feature vectors, and then apply the denoising algorithm to denoise the ultrasonic signal, further reduce the noise level in the signal, improve the signal purity, and retain the main characteristics and useful information of the signal; Step 5: Signal enhancement: Based on denoising, the signal is further enhanced and restored through the signal enhancement algorithm to enhance the effective components of the signal, improve the signal-to-noise ratio, and improve the signal's amplitude, strength, clarity, contrast, anti-interference ability, transmission efficiency, detection ability, amplitude and power to ensure the purity and stability of the signal; Step 6: Data compression and reconstruction: Data compression is performed on the denoised and enhanced ultrasonic signal to reduce the burden of data storage and transmission, improve the efficiency of data processing, help reduce the operating cost of the system, and improve the overall performance of the system; the data compression algorithm through lossy compression or lossless compression should be able to retain the main features and useful information of the signal while reducing the redundancy and complexity of the data; Reconstruct the compressed data according to the feature vector to restore the original information and detailed features of the signal, ensuring that the reconstructed signal is not distorted and maintains high position accuracy; accurately restore the waveform and characteristic parameters of the signal through the data reconstruction algorithm, while avoiding the introduction of new noise and interference, maintaining the accuracy and reliability of the signal; output ultra-low noise ultrasonic signals; Step 7: Signal optimization and output: The reconstructed signal is further optimized, including but not limited to smoothing and artifact removal, to reduce burrs and artifacts in the signal, enhance the signal strength through algorithms, improve the signal-to-noise ratio, further reduce noise interference and improve signal accuracy; Output the optimized signal to the subsequent processing system or application equipment. The output format can be selected according to specific needs. The output format includes but is not limited to digital signals and analog signals. This step is the end of the signal processing flow. It is necessary to ensure that the quality and stability of the output signal meet the requirements of subsequent processing. Step 8, post-processing and result analysis: smooth the processed signal to remove possible artifacts or outliers; perform data normalization or standardization for subsequent analysis and application; Performing quality assessment on the processed signal, where the quality assessment categories include but are not limited to the degree of improvement in signal-to-noise ratio and degree of distortion; Use visualization tools to display and analyze signals. Visualization categories include but are not limited to time domain waveforms and spectrum graphs. Step 9: Method optimization and iteration: Algorithm optimization: Based on actual application scenarios and needs, combined with artificial intelligence and deep learning modules, evaluate the performance of the algorithm, including but not limited to denoising effect, signal enhancement effect and data compression efficiency, and use comparative experiments and statistical analysis methods for continuous evaluation and adjustment; Adjust the feature vector extraction method and denoising algorithm design parameters based on the processing results, optimize and iterate the feature extraction, noise estimation and denoising algorithms. The system can adaptively adjust the processing parameters to improve the accuracy, performance and efficiency of signal processing and reduce computational complexity and resource consumption. System integration and testing: Integrate the ultra-low noise signal processing method into the ultrasonic signal processing system, perform system testing and verification, and ensure the stability and reliability of the system.

Images